venue: SoCC
year: 2024
date: "2024-11-01"
tags: [cloud-computing, distributed-systems]

selected:
  - dblp_key: "conf/cloud/PatkeRJQPNKI24"
    title: "Queue Management for SLO-Oriented Large Language Model Serving"
    tldr: "A queue management framework that enforces latency SLOs for LLM serving by dynamically routing and prioritizing requests across heterogeneous inference capacity."
    why_notable: "As LLM deployments move into production clouds, meeting strict time-to-first-token and total latency SLOs becomes critical; this work directly addresses that gap with a practical, deployable solution. It is one of the first papers to treat LLM serving as a cloud SLO-management problem rather than a pure model-optimization problem."

  - dblp_key: "conf/cloud/LiuYOJY0HHSZWH24"
    title: "Kale: Elastic GPU Scheduling for Online DL Model Training"
    tldr: "Kale elastically resizes GPU allocations for online DL training jobs in response to real-time resource pressure, improving cluster utilization without violating training progress guarantees."
    why_notable: "Elastic GPU scheduling is an unsolved pain point in shared ML clusters; Kale's ability to dynamically shrink and expand jobs without checkpointing overhead is directly applicable to production training infrastructure at hyperscalers."

  - dblp_key: "conf/cloud/WangWZ24"
    title: "Hops: Fine-grained heterogeneous sensing, efficient and fair Deep Learning cluster scheduling system"
    tldr: "Hops uses fine-grained, heterogeneity-aware GPU sensing to make scheduling decisions that are simultaneously efficient and fair across diverse DL workloads."
    why_notable: "Hardware heterogeneity in GPU clusters is the norm, not the exception; Hops provides a principled framework for exploiting that diversity, making it immediately relevant to operators of mixed-generation GPU fleets."

  - dblp_key: "conf/cloud/CopikCRBBH24"
    title: "Process-as-a-Service: Unifying Elastic and Stateful Clouds with Serverless Processes"
    tldr: "PraaS introduces a long-lived, stateful serverless process abstraction that bridges the gap between ephemeral FaaS functions and persistent cloud VMs."
    why_notable: "Statelessness is the central limitation of today's FaaS platforms; this paper proposes a well-grounded new programming model that could reshape how developers think about serverless, backed by implementation and evaluation at scale."

  - dblp_key: "conf/cloud/ZhaoYWDQ024"
    title: "FaPES: Enabling Efficient Elastic Scaling for Serverless Machine Learning Platforms"
    tldr: "FaPES achieves fast, fine-grained vertical and horizontal scaling of serverless ML serving pods by decoupling memory provisioning from compute allocation."
    why_notable: "Elastic scaling for ML inference is a key cost driver in cloud ML platforms; FaPES demonstrates sub-second scaling decisions that reduce both cold-start overhead and resource waste, with results from a production deployment."

  - dblp_key: "conf/cloud/ZhangH0024"
    title: "Faascale: Scaling MicroVM Vertically for Serverless Computing with Memory Elasticity"
    tldr: "Faascale enables runtime vertical memory scaling of Firecracker microVMs for serverless functions, eliminating the need to restart or pre-provision fixed memory sizes."
    why_notable: "Memory over-provisioning is a major cost inefficiency in serverless platforms; Faascale's live memory elasticity directly reduces waste while maintaining the isolation guarantees of microVM-based FaaS."

  - dblp_key: "conf/cloud/HasanZU24"
    title: "AutoBurst: Autoscaling Burstable Instances for Cost-effective Latency SLOs"
    tldr: "AutoBurst exploits burstable cloud instance types and their CPU credit mechanics to autoscale services at lower cost while still meeting tail-latency SLOs."
    why_notable: "Burstable instances are widely available on all major clouds yet poorly understood for SLO-sensitive workloads; this paper provides a rigorous autoscaling policy that unlocks significant cost savings without sacrificing latency guarantees."

  - dblp_key: "conf/cloud/GuptaNHCTSRLWWG24"
    title: "Dynamic Idle Resource Leasing To Safely Oversubscribe Capacity At Meta"
    tldr: "Meta's production system dynamically lends idle reserved-capacity to opportunistic workloads, recovering stranded compute while ensuring low-latency eviction when owners reclaim resources."
    why_notable: "This industry paper provides rare visibility into hyperscale capacity management at Meta's scale, demonstrating that safe oversubscription can recover tens of percent of otherwise idle datacenter capacity."

  - dblp_key: "conf/cloud/DiaoHKSBDPPNNK24"
    title: "Forecasting Algorithms for Intelligent Resource Scaling: An Experimental Analysis"
    tldr: "A comprehensive empirical study comparing classical and learned forecasting algorithms for cloud autoscaling, yielding concrete guidelines on when each approach wins."
    why_notable: "Autoscaling relies heavily on workload forecasting, yet practitioners lack principled guidance on algorithm choice; this work from the MIT/AWS group fills that gap with rigorous experimentation across real-world cloud traces."

  - dblp_key: "conf/cloud/SinghSNVK24"
    title: "Vista: Machine Learning based Database Performance Troubleshooting Framework in Amazon RDS"
    tldr: "Vista is a production ML framework deployed in Amazon RDS that automatically diagnoses performance regressions by correlating database metrics with causal performance models."
    why_notable: "Database performance debugging at cloud scale is labor-intensive and error-prone; Vista's deployment in RDS demonstrates how ML-driven root-cause analysis can reduce mean-time-to-resolution for thousands of customer instances."

  - dblp_key: "conf/cloud/AgrawalRBNV0T24"
    title: "Inshrinkerator: Compressing Deep Learning Training Checkpoints via Dynamic Quantization"
    tldr: "Inshrinkerator applies dynamic quantization to DL training checkpoints at save time, reducing checkpoint sizes by up to 4x with negligible impact on training convergence."
    why_notable: "Checkpoint storage and I/O are significant costs in large-scale distributed training; this work provides a transparent, easy-to-adopt compression layer that can be retrofitted into existing training pipelines."

  - dblp_key: "conf/cloud/BashirGSSIOD24"
    title: "The Sunk Carbon Fallacy: Rethinking Carbon Footprint Metrics for Effective Carbon-Aware Scheduling"
    tldr: "The paper argues that conventional carbon metrics misattribute embodied (manufacturing) carbon as a fixed sunk cost, and proposes revised metrics that make carbon-aware scheduling decisions more accurate and actionable."
    why_notable: "Carbon-aware cloud scheduling is an emerging priority, but flawed metrics can lead to counterproductive decisions; this work from the Delimitrou and Irwin groups provides a conceptual correction with broad implications for green cloud policy and tooling."